We have made a comparative analysis of the extracellular phospholipid composition of biofilms of Gram-negative and Gram-positive bacteria. The surface film of a growing bacterial community contains small membrane vesicles and a bilipid layer covering the entire surface of that community. These supracommunity films containing the bilipid layer can cover the entire surface of a Petri dish and form a boundary between bacterial communities and the environment. A mixed bacterial lawn, formed by unrelated bacteria, also becomes covered with a single film containing a lipid bilayer. The phospholipid compositions of the bacterial cell and surface film bilipid layer reflect the nature of the bacterial strains forming the community, but have some specific differences.

This special issue of Biofilms highlights current experimental studies and modeling of collective phenomena in single-celled microbes, primarily bacteria, stemming from the Workshop “Biocomplexity VI: Complex Behavior in Unicellular Organisms”, held 12–16 May 2004, at Indiana University, Bloomington, IN, USA, and co-organized by the Biocomplexity Institute at Indiana University and the Interdisciplinary Center for the Study of Biocomplexity at the University of Nôtre Dame, with support from the National Science Foundation (grant no. 0352904), the National Institutes of Health (grant no. GM071709-01), the National Institute of General Medical Sciences and the College of Arts and Sciences at Indiana University.

The concept that bacteria live preferentially in matrix-enclosed communities attached to surfaces has emerged gradually from scientific observations over an extended period of time, but the pace at which this concept has advanced has accelerated sharply during the past two decades. Because Antonie van Leuwenhoek examined his own teeth scrapings with his primitive microscope, he probably saw more biofilm fragments than planktonic cells, and dental microbiologists and waste-water engineers have had a lengthy association with biofilms without using that term. Early microscopic observations of marine systems showed that most bacteria adhered actively to surfaces, and the role of surfaces in the migration and maturation of myxobacterial communities was noted very early in the study of these fascinating organisms. The new concept that was crystallized in a Scientific American article in February 1978 (Costerton, J. W., Geesey, G. G. & Cheng, K. J. (1978) How bacteria stick. Scientific American238, 86–95) was that these surface associations were the rule (and not the exception) in all nutrient-sufficient microbial ecosystems, and that most bacteria in the biosphere grow in biofilms.

Biofilms colonizing ancient limestone Mayan monuments in Uxmal (Yucatán, Mexico) were characterized for their microbial composition and differences using phospholipid fatty acid (PLFA) analysis and single-strand conformation polymorphism (SSCP) of polymerase chain reaction (PCR)-amplified genes of the rRNA small subunit. Biofilms 1–7, displaying different macroscopic characteristics, were drawn from interior and exterior limestone walls, which were characterized by different microenvironmental conditions. Inner surfaces that were exposed to low light intensity and higher water availability supported a higher biofilm biomass, probably indicating that these environmental conditions are more suitable for biofilm growth and development. Sequencing of SSCP profiles from the biofilms showed that bacteria affiliated with the phyla Proteobacteria, Bacteroidetes and Cyanobacteria colonized both internal and external surfaces, although certain Nevskia and Salinibacter were particularly associated with internal biofilms. In contrast, a particular biofilm community composed almost exclusively of Rubrobacter-related bacteria was found in only one exterior partially shaded sample (biofilm 3), characterized by high illumination and low water content. No specific organisms were detected in biofilm 7. The PLFA profiles suggested that cyanobacteria were the biomass-dominant group in most biofilms, except for biofilm 3 (exterior) where actinobacterial markers were detected in significant proportions. Interestingly, most of the detected sequences were related to halophylic bacteria, although the similarity of these clones to known sequences in databases was low (< 94%). This finding suggests that much bacterial novelty, probably of halophylic nature, remains to be identified in these biofilms. Our study suggests that water availability and light regime appear to be the main environmental determinants defining boundaries for biofilm formation, while substratum salinity appears to be an important abiotic factor that influences biofilm community structure in these specialized microbial habitats.

Pseudomonas aeruginosa PAO1 was grown on 40° ZnSe and 40° AMTIR® (Amorphous Material Transmitting Infrared Radiation, Ge33As12Se55) internal reflection elements in a thermostatted flow cell. Information about the chemistry of the biofilm development was obtained using Fourier transform infrared spectroscopy. The weakly absorbing C–H stretching region of the infrared spectrum was measured for the first time and revealed qualitative and quantitative changes not evident in the amide region studied by previous investigators. There is a chemical difference between the early growth period and subsequent development of the biofilm.

Biofilms are often considered as localized zones of high cell density. Quorum sensing provides a means for control of population processes and has been implicated in the regulation of biofilm activities. We present a role for quorum sensing in programmed detachment and dispersal processes. Biofilms of Pseudomonas aeruginosa PAO1 and its isogenic homoserine lactone (HSL) mutant P. aeruginosa PAO-JP2 were grown in batch culture on glass substrata; differences were found in the rate and extent of formation of biofilm. Climax communities were observed for PAO1 at 24 h. These were later accompanied by foaming, a drop in the surface tension of culture media and dispersal of the biofilm, after which no subsequent biofilm accretion occurred. PAO-JP2 cultures reformed biofilm post-detachment and did not foam. Prevention of biofilm reformation in the wild type was related to some component excreted into the culture medium. Rhamnolipid, a biosurfactant regulated by quorum sensing, was detected in PAO1 cultures. When rhamnolipid was added to freshly inoculated substrata, biofilm formation was inhibited. At 20 h, PAO1 biofilms were transferred to medium with added rhamnolipid: biofilm was relatively unaffected. Biofilm events were also studied in medium supplemented with N-butyryl-L-homoserine lactone, which is involved in the regulation of rhamnolipid synthesis. Both strains exhibited similar trends of rapid biofilm formation and dramatic changes in the rate and extent of biofilm accretion. In both cases, there was premature foaming, lowered surface tension and elevated rhamnolipid levels. A role for HSLs in maintenance of biofilm and events leading to dispersion of cells is proposed. This role would encompass dispersion but not necessarily detachment of cells from biofilm and supports a new function for rhamnolipid in pathogenesis, whereby rhamnolipid would promote the dissemination of cells from a nidus of infection.

The aim of this research was to study the effect of varying incubation times and sucrose exposures on lesion development and biofilm composition using a multi-species biofilm caries model. Two studies were conducted. In study 1, enamel specimens were divided into four groups, inoculated with a mixed overnight culture of Streptococcus mutans, Lactobacillus casei, Actinomycesnaeslundii, Streptococcusparasanguis and Streptococcussalivarius, and exposed to circulating trypticase soy broth + 5% (w/v) sucrose (TSBS; 30 min, three times per day (3 × /day)) and a mineral washing solution containing 0.25 p.p.m. fluoride (22.5 h/day) for 2, 5, 6 or 8 days. In study 2, additional enamel specimens were divided into four groups and exposed to the same biofilm for 7 days, but with variations in the feeding schedule: TSBS 3×/day for 5 min, TSBS 3×/day for 15 min, TSBS 3 × /day for 30 min, or TSBS 3×/day for 30 min + 1×/day for 15 min. At the end of each study, bacterial colonization counts and lesion size were determined. In study 1, specimens developed significantly deeper carious lesions with longer demineralization time (average lesion depth was 52.16 μm, 67.86 μm, 84.91 μm, and 99.97 μm, respectively, for 2, 5, 6, and 8 days). In study 2, there was no significant difference in size between the lesions developed at feeding schedules of 3 × /day for 5 or 15 min. Lesions exposed to longer (30 min) and more frequent feeding schedules (4×/day) were significantly larger than the other groups. In both studies, all five bacterial species were able to colonize the enamel and were present in all groups at the end of the experiments, with predominance of lactobacilli over S. mutans. In conclusion, larger lesions were observed with increased incubation time and more frequent feeding schedules, with small variations in biofilm composition.

Oral bacteria form mixed-species biofilms known as dental plaque. Growth of these complex microbial communities is often controlled with the use of antimicrobial mouthrinses. Novel laboratory methods for testing the efficacy of antimicrobials in situ are necessary to complement current clinical testing protocols. In this study, we examined the effects of antimicrobial agents on a streptococcal biofilm grown in a saliva-conditioned flowcell. The flowcell coupled with confocal laser scanning microscopy enabled examination of growing oral biofilms in situ without disruption of the microbial community. Biofilms composed of Streptococcus gordonii DL1 were grown in an in vitro flowcell and treated with several commercially available antimicrobial mouthrinses containing essential oils, triclosan, cetylpyridinium chloride/domiphen or chlorhexidine. The results of this study revealed varying abilities of the antimicrobial agents to cause cellular damage on the growing biofilm in situ. This study therefore demonstrated the usefulness of the flowcell in the rapid assessment of antimicrobial efficacy.

A fermentation system was designed to model the human colonic microflora in vitro. The system provided a framework of mucin beads to encourage the adhesion of bacteria, which was encased within a dialysis membrane. The void between the beads was inoculated with faeces from human donors. Water and metabolites were removed from the fermentation by osmosis using a solution of polyethylene glycol (PEG). The system was concomitantly inoculated alongside a conventional single-stage chemostat. Three fermentations were carried out using inocula from three healthy human donors.

Bacterial populations from the chemostat and biofilm system were enumerated using fluorescence in situ hybridization. The culture fluid was also analysed for its short-chain fatty acid (SCFA) content. A higher cell density was achieved in the biofilm fermentation system (taking into account the contribution made by the bead-associated bacteria) as compared with the chemostat, owing to the removal of water and metabolites. Evaluation of the bacterial populations revealed that the biofilm system was able to support two distinct groups of bacteria: bacteria growing in association with the mucin beads and planktonic bacteria in the culture fluid. Furthermore, distinct differences were observed between populations in the biofilm fermenter system and the chemostat, with the former supporting higher populations of clostridia and Escherichia coli. SCFA levels were lower in the biofilm system than in the chemostat, as in the former they were removed via the osmotic effect of the PEG. These experiments demonstrated the potential usefulness of the biofilm system for investigating the complexity of the human colonic microflora and the contribution made by sessile bacterial populations.

Biofilm-formation in Staphylococcus epidermidis depends on the expression of the icaADBC operon encoding the enzymes required for the synthesis of polysaccharide intercellular adhesin (PIA). Different S. epidermidis strains vary widely in the degree of PIA and biofilm that they produce. In 11 clinical S. epidermidis strains we analyzed the biofilm-forming capacity in relation to the amount of ica expressed in static biofilm cultures. In mid-exponential growth phase no correlation could be detected between the level of ica transcription and the biofilm-forming phenotype. When the different strains were grown under conditions leading to a biofilm-negative phenotype, ica-expression was highly upregulated. Sequence analysis demonstrated that the observed differences were not due to major mutations in the ica promoter region but apparently to other strain-specific regulators.

The study of bacterial colonies is a crucial step towards understanding biofilms. We review some of the exciting discoveries about bacterial self-organization that might shed new light on biocomplexity in general and biofilms in particular. This review is aimed at researchers from different disciplines – microbiology, biology, chemistry, physics, mathematics and computer science. To make the presentation comprehensible we have avoided the use of specialized terminology of the different disciplines and limited the experimental and computational details.

Bacteria can self-organize into hierarchically structured colonies of 109 to 1012 bacteria, each utilizing a great variety of biochemical communication agents, such as simple molecules, polymers, peptides, complex proteins, genetic material and also “cassettes of genetic information” such as plasmids and viruses. Bacteria use their intracellular flexibility, involving signal transduction networks and genomic plasticity, to collectively maintain self and shared interpretations of chemical cues, exchange of meaning-bearing chemical messages, and dialogues. The meaning-based communication permits the formation of colonial intentional behavior, purposeful alteration of colony structure and decision-making – features we might begin to associate with bacterial social intelligence. Such social intelligence, should it exist, would require going beyond communication to encompass additional intracellular processes, as yet unknown, for generating inheritable colonial memory and commonly shared genomic context.

There is growing evidence to suggest that non-typeable Hemophilus influenzae (NTHi), an important cause of otitis media in children, is able to grow as a biofilm in the middle ear. This observation may help to explain bacterial persistence in chronic infections. We evaluate the usefulness of rapid freezing and freeze substitution as a means of preparing biofilms for ultrastructural examination by comparing the morphology of cryofixed specimens with the morphology produced using more conventional chemical fixation and dehydration methods. Chemical fixation and dehydration methods produced substantial ultrastructural damage to individual NTHi in the biofilm and loss of extracellular matrix, even in the presence of ruthenium red. In comparison, cryofixed and freeze-substituted NTHi biofilms showed significantly improved preservation of bacterial ultrastructure and biofilm organization. The intracellular contents of NTHi prepared using the cryomethods showed little evidence of aggregation, and bacteria within the biofilm were closely packed and surrounded by an abundant extracellular matrix. Although high-pressure freezing of NTHi biofilms followed by freeze substitution was highly effective for preserving ultrastructure when examined by transmission electron microscopy, immersion in liquid propane offered an alternative, “less technical”, freezing method. Immersion in liquid propane followed by freeze substitution and critical point drying was most effective for preserving ultrastructural details in specimens examined by scanning electron microscopy.

A natural microphytobenthic assemblage from the Eden Estuary, Scotland, was used to study the effect of temperature and irradiance on the sustainability and species composition of a natural transient biofilm. Three tidal tanks were maintained: tank 1 at 10 °C; tank 2 at 18 °C; and tank 3 at 26 °C. Within each tank, five cores were unshaded (350 μmol/m2 per s), five cores were semi-shaded (175 μmol/m2 per s), and five cores were shaded (70 μmol/m2 per s). Chlorophyll a increased in all treatments, but accumulated slower with increasing temperature. Surface diatom biomass of biofilms (as determined from measurements of minimum fluorescence, Fo15), grown at 10 °C and 18 °C was sustained above initial levels after 21 days, whilst biofilms grown at 26 °C suffered a severe loss of diatoms after 14 days, probably owing to nutrient limitation. Species richness and diversity of diatom assemblages illustrated a variety of responses to the temperature and light conditions. Diatoms at 10 °C acclimated to the different light levels by varying the ratios of diadinoxanthin: chlorophyll a, whilst at 18 °C the diatom species composition changed dramatically in response to shading. Ratios of zeaxanthin: chlorophyll a increased with increasing temperature, indicating an increase in cyanobacterial biomass at 26 °C after 21 days. Increased temperature significantly increased the maximum theoretical electron transport rate (rETRmax) in the short term (days), although light and temperature treatments did not affect the maximum light utilization coefficient (αrETR). Limitations of the fluorescence methodology used to study the resultant mixed community of benthic phototrophs is discussed. Diatom vertical migration in response to light, and an alteration of the main functional taxonomic group, has implications for the interpretation and value of fluorescence data.

Listeria monocytogenes is ubiquitous in nature and a major concern for the food industry, since it is the causal agent of the serious foodborne illness listeriosis. This organism can be introduced through many routes to food-processing environments and may become established on food-processing equipment. Subsequently, food products may become contaminated during processing. In addition, the bacterium can grow at refrigeration temperatures. Biofilms are regarded as important with respect to the survival and growth of microorganisms in the food industry. Microorganisms growing in biofilms are protected against cleaning and disinfection and are difficult to eradicate. Listeria monocytogenes may grow in biofilms that protect them against environmental stress and can be isolated from surfaces after cleaning and disinfection. For each individual food-processing plant, a limited number of clones of L. monocytogenes may become established and persist for years. Persistent strains adhere to surfaces and form biofilms more readily compared to sporadically found strains, suggesting that adherence to surfaces is important for survival and persistence of L. monocytogenes in food-processing environments. Listeria monocytogenes can adhere to all the materials commonly used in the food industry. In biofilms L. monocytogenes is significantly more resistant to disinfection than its free-living counterparts and thick, complex biofilms are more difficult to remove than adherent single cells of L. monocytogenes. Several novel approaches to avoid adhesion of L. monocytogenes have been proposed, but high costs, practical difficulties or resistance problems limit their practical use. Despite considerable research on the adhesive properties and resistance of L. monocytogenes enabling its survival in the food production environment, a final solution for avoiding establishment of the bacterium has not yet been found.

Jamming the command language of bacteria: a new approach to the control of bacterial infections

The cep quorum-sensing system of Burkholderia cepacia H111 plays a pivotal role for biofilm formation and pathogenicity

Detachment of Shewanella oneidensis MR1 biofilms

Protozoan grazing on biofilms and the role of quorum sensing in grazing resistance

Control of heat exchanger biofilms

Computer-aided design of a novel antibacterial agent for the treatment of Burkholderia cepacia infections

Anaerobic regulatory network essential for efficient Pseudomonas aeruginosa biofilm formation

Development and testing of a new phosphorylcholine polymer coating to prevent biofilm formation on optical fiber sensor systems

Bactericidal layers based on silver nanoparticles

Metagenomics: a novel tool to exploit microbial communities for industrial applications

Formation of biofilms – do not forget the hydrodynamics

Microbial adhesion – a target for the development of household and industrial products

Concept of a specialty paper producer to avoid microbiological problems

Identification of primary biofilm organisms on steel surfaces of bottling plants

Carbon sharing in complex microbial biofilms grown on polychlorinated biphenyl droplets

Development of methods for assessing microbial colonization of coil-coated steel sheets

Microbial in situ activity assessment using “BACTRAPS” with 13C-labelled substrates

Magnetic resonance imaging as a tool for non-invasive imaging of biofilms

Phototrophic biofilms – development and growth characteristics in a special flow chamber

A new mechanism of bacterial adhesion to materials and organisms containing the chitin and/or chitosan

Extracellular polymeric substances – the “glue” of biofilms

Sphingomonas biofilm architecture grown under different hydrodynamic conditions

Molecular basis of biofilm formation in staphylococci and its implication for virulence

Extracellular polymeric substances – the dark matter of biofilms

The multi-functional Staphylococcus aureus autolysin Aaa mediates adherence to immobilized fibrinogen and fibronectin

Hierarchy within the regulatory cascade of Staphylococcus epidermidis biofilm formation

Human biofilms in situ

An in vitro testing system based on bioreactor technology for the analysis of biofilm formation on dental materials

Immune compounds – inhibitors of microbial adhesion in cell culture

A proteomic approach to identify quorum-sensing regulated and surface-induced proteins of the rhizobacterium Pseudomonas putida IsoF

Evidence that IcaB deacetylases the staphylococcal N-acetylglucosamine polymer polysaccharide intercellular adhesin

Microelectrode and confocal laser scanning microscopy study of particle-supported biofilms

Biocidal silica coatings

Inhibition of adherence of coagulase-negative staphylococci to acrylic by subinhibitory concentrations of antibiotics

The influence of antiplaque agents on streptococcal biofilm grown on enamel and glass

Comparison of digital image analysis programs for confocal laser scanning microscopy data sets of biofilms

Fungal biofilms on mineral materials

The influence of subinhibitory concentrations of fluconazol and amphotericin B in the adhesion of different Candida species to acrylic

Inhibition of biofilm formation by mutants group streptococci on toothbrush bristles in vivo

Quorum sensing of biofilm-associated Erwinia persicinus

Growth with sodium dodecyl sulphate induces aggregation and biofilm formation of Pseudomonas aeruginosa strain PAO1

Microbial contamination of stationary eyewash units in laboratories

Microbial biofilms on building stone

Scanning transmission X-ray mapping of microbial biofilms and associated metals

A marine biofilm formation model in multi-well plates

Secondary metabolites in bacterial biofilms and their protective function against protozoan grazing

Analysis of Deinococcus geothermalis biofilm properties with lectin staining and confocal laser scanning microscopy

Effect of the surface roughness of cast titanium on bacterial colonization

Metabolic activity of microbial biofilm on hydroxyapatite-coated titanium surfaces

Multiple quorum sensing in hypogean biofilms

Determination of adsorption isotherms for the attachment of Deinococcus geothermalis by an enzyme-activity-based method

Biosurfactant from Streptococcus thermophilus A inhibit microbial adhesion on silicone rubber

The influence of nitrogen on the development of biofilms

Biogeography of biliary stent community biofilms

Development of a novel screening system for the characterization of biofilm growth on different carriers

Prevention of Pseudomonas fluorescens adhesion to surfaces using bacteriophage S1

ConAn, a new software developed for digital image analysis of three-dimensional biofilm data

Lectin-binding-analysis of rotating annular reactor cultivated biofilms

Adhesiveness and hydrophobicity of food spoilage yeasts

Investigation of microbial adhesion on a thickness shear mode resonator

Staphylococcus epidermis adhesion on modified urea/urethane elastomers

Pseudomonas aeruginosa lectin LecB is a cell surface protein involved in biofilm formation

Aerobic biofilms as Trojan horses for anaerobic product spoiling bacteria?

Reduced fouling tendencies of ultrafiltration membranes in waste-water treatment by plasma modification

A universal method for analysing the diversity of mycosporines and mycosporine-like amino acids in microorganisms and environmental rock inhabiting microbial communities

In vivo biofilm formation on dental hard tissues

Implant-associated osteomyelitis: analysis of the local cellular immune response

Fluorescence lifetime imaging (FLIM) of interfacial microbial communities

Concepts of electrochemical treatment and useful techniques for proper implementation

Inhibition of Staphylococcus aureus biofilm formation on vascular graft surfaces coated with elemental silver

Planktonic versus biofilm-grown Staphylococcus aureus cells – determination of different expression patterns with DNA microarrays

Diversity of the microbial community and the catabolic genes in biofilms grown on polychlorinated biphenyls droplets

A modular approach to the mathematical modelling of biofilm systems

The study of castor-oil-derived polyurethane in relation to biofilm

Metagenome survey of biofilms in drinking water networks

High cell density and close proximity of diverse species of microorganisms are typical of life in natural biofilms. These conditions give ample opportunity for both competitive and cooperative interactions between individuals of the same and different species. Cooperative behaviour benefits the group of neighbouring microbes but comes at a fitness cost for the cooperating individuals. This creates a conflict of interest between the fitness of the individual and the fitness of the group. Individuals that defect from cooperation and therefore do not pay the cost but nevertheless benefit from the cooperative behaviour of others are called cheaters. Cooperative behaviour in the presence of cheaters constitutes altruism towards the cheaters. The aim of this review is two-fold: first, to introduce key concepts from kin selection and group selection theory that allow us to understand how cooperative behaviour can evolve in the face of cheaters; secondly, to draw attention to the conflicts of interest prevalent in biofilms yet largely ignored in the biofilm literature. Examples discussed comprise growth restraint in stationary phase as an instance of the Prisoner's Dilemma, growth restraint to allow channel formation, restraint in resource consumption or economical use of resources as altruistic behaviour, population heterogeneity as insurance against environmental changes, cooperative investment in diffusible exoenzymes, cooperation of pathogens and virulence, diffusion sensing versus quorum sensing and the inflation of signals, antibiotic resistance as collective action, and programmed cell death.

Tolerance of Staphylococcus epidermidis in biofilms to killing by rifampin was correlated with limitation of bacterial growth in the biofilm state. Intact biofilm experienced a 0.62 log reduction when treated with 0.1 μg rifampin/ml for 4 h whereas the same treatment of exponential-phase planktonic cells produced a log reduction of 4.48. Stationary-phase planktonic cells were nearly as tolerant as intact biofilm cells, experiencing a 1.11 log reduction. Biofilm bacteria grew at only 10% of the maximum rate at which they grew on the same medium in planktonic culture. Killing was localized near the surface of the biofilm adjacent to the nutrient source, as revealed by staining with a respiratory dye. Increased nutrient concentration during antibiotic treatment enhanced killing of biofilm cells. Changing the oxygen tension in the gas phase above the biofilm during antibiotic treatment barely affected killing. It was hypothesized that the biofilm harbors significant numbers of stationary-phase-like cells in the nutrient-limited depths of the biofilm, and that these inactive cells are the ones that survive antibiotic challenge.

Microbial biofilm formation can be influenced by many physiological and genetic factors. The conventional microtiter plate assay provides useful but limited information about biofilm formation. With the fast expansion of the biofilm research field, there are urgent needs for more informative techniques to quantify the major parameters of a biofilm, such as adhesive strength and total biomass. It would be even more ideal if these measurements could be conducted in a real-time, non-invasive manner. In this study, we used quartz crystal microbalance (QCM) and microjet impingement (MJI) to measure total biomass and adhesive strength, respectively, of S. mutans biofilms formed under different sucrose concentrations. In conjunction with confocal laser scanning microscopy (CLSM) and the COMSTAT software, we show that sucrose concentration affects the biofilm strength, total biomass, and architecture in both qualitative and quantitative manners. Our data correlate well with previous observations about the effect of sucrose on the adherence of S. mutans to the tooth surface, and demonstrate that QCM is a useful tool for studying the kinetics of biofilm formation in real time and that MJI is a sensitive, easy-to-use device to measure the adhesive strength of a biofilm.

The aim of this study was to investigate bacterial survival and biofilm formation on toothbrushes. Fifteen healthy volunteers each used a normal toothbrush and an antibacterial toothbrush of the same design for two separate 5 week periods. Bacteria were removed from the brush head by swabbing and mechanical agitation in 10ml of tryptone soya broth, cultured aerobically on selective and non-selective media, and classified by Gram staining, catalase and oxidase tests. Survival of Staphylococcus epidermidis and Pseudomonas aeruginosa was monitored in the laboratory on both types of brush over 8 days. Scanning electron microscopy was used to observe biofilm formation on antibacterial and conventional brushes used for various times. Numbers of bacteria isolated from conventional and antibacterial brushes from different individuals ranged from 8.3×103 to 4.7×106 and from 1×102 to 1.2×106 colony-forming units/ml, respectively. A larger number of bacteria were isolated from conventional brushes than from antibacterial brushes used by the same individuals but no statistically significant difference was demonstrated. No differences in the relative proportions of Gram-negative and Gram-positive rods or cocci were seen. Staphylococci, presumptive coliforms and pseudomonads were isolated from 48%, 28% and 16% of brushes, respectively. Pseudomonas aeruginosa was viable for at least 4 days on conventional, and 2–3 days on antibacterial, brushes, whilst S. epidermidis survived for 6–8 days on antibacterial and more than 8 days on conventional brushes. Biofilms formed on the heads and bristles of both conventional and antibacterial brushes. Extensive, mixed community biofilms developed after several months of use. We conclude that toothbrushes may be a reservoir of opportunistic pathogens including staphylococci and pseudomonad-like organisms and must be considered as a potential source of haematogenous infections and cross-infection.